Research On Coverage Optimization For WSN Based On Swarm Intelligence Algorithm

Wireless sensor network(WSN),as the critical technology of long-distance regional monitoring system,can provide efficient sensing and communication services under limited energy supply.Coverage control is an important method to ensure efficient communication and reliable data transmission.Given that the complex physical environment constrains the deployment of nodes and hinders energy replenishment and recovery,our research motivation is to repair coverage holes and reduce energy consumption during the process of sensor node redeployment,so as to optimize and enhance the coverage of WSN.In recent years,the advancement and maturity of swarm intelligence algorithm has facilitated its application in WSN area coverage enhancement,and many research results have adopted this kind of algorithm to realize the redeployment of sensor nodes.Therefore,this paper focuses on optimizing the node deployment of homogeneous and heterogeneous WSN by using swarm intelligence algorithm in two-dimensional plane environment,and puts forward two coverage control strategies,aiming at making WSN better serve users.The research work and achievements of this paper are as follows:(1)In the homogeneous WSN node deployment environment,taking the network coverage rate as the optimization goal,a virtual force-directed improved moth flame optimization algorithm(VF-IMFO)is designed in this paper.By combining virtual force with improved moth flame algorithm,the phenomenon of covering holes and redundant coverage in initial random deployment of nodes is overcome.Firstly,the spiral shape is dynamically adjusted by changing the spiral position updating strategy to enhance the search performance of moths in unknown space.In order to expand the information dimension of moth search behavior,an adaptive inertia weight strategy considering the historical optimal flame average value is introduced.Secondly,a covering disturbance strategy based on virtual force is proposed,which makes up for the stagnation of moth individual optimization.By analyzing the virtual force of nodes in the network,the resultant force is calculated,the nodes’ position is optimized based on the resulting force,and rate of global convergence was expedited to avoid the precocious phenomenon.In addition,with the aim of avoiding the fluctuation of coverage rate in the late iteration,the virtual moving step size is set to a parameter related to the number of iterations,for algorithm stability.(2)In heterogeneous WSN node deployment environment,the coverage optimization problem is constructed as a multi-objective problem by jointly optimizing the network coverage rate and average distance moved by the nodes,exploiting the idea of Pareto optimal solution sets,this paper designs a non-dominated sorting multi-objective tuna swarm optimization algorithm(NSMTSO).To enhance the diversity of the tuna population,Sobol sequence was utilized for initialization in the initial stage.Additionally,non-linear weight coefficients were introduced to comprehensively consider the global search and local exploitation abilities of the algorithm.The fast non-dominated sorting method in NSGA-II is used for reference to classify individuals of tuna group,and elite reservation and external archiving strategies are added at the same time.Moreover,a reasonable crowding distance calculation method was proposed to enhance the diversity of the optimal solutions and improve the quality of the Pareto solution set.(3)Conduct experiments in Matlab simulation environment to compare proposed algorithm with other excellent algorithms on standard test functions and WSN coverage optimization.The simulation suggests that IMFO algorithm is more effective than primal MFO in addressing issues related to slow convergence and restricted search capabilities.In different WSN deployment environments,the VF-IMFO algorithm maintains good coverage rate,and its optimised node deployment is more balanced.Furthermore,the NSMTSO algorithm exhibits greater stability and produces a more competitive set of optimal solutions in terms of convergence and distribution.An effective trade-off between the two objectives of network coverage rate and average distance between nodes enables decision makers to find the best solution for deploying WSN nodes.